| 时间: 2022-01-10 | 次数: |
段铭钰, 袁瑞甫, 杨艺.基于改进RBF神经网络的采煤机截割煤岩性状智能识别[J].河南理工大学学报（自然科学版）,2022,41(1):43-51.
DUAN M Y, YUAN R F, YANG Y. Intelligent recognition of coal and rock properties in shearer cutting process based on improved RBF neural network [J].Journal of Henan Polytechnic University(Natural Science) ,2022,41(1):43-51.
基于改进RBF神经网络的采煤机截割煤岩性状智能识别
段铭钰1,2, 袁瑞甫3, 杨艺4
1.华中科技大学 机械科学与工程学院，湖北 武汉 430070；2.河南大有能源股份有限公司，河南 三门峡 472300；3.河南理工大学 能源科学与工程学院，河南 焦作 454000；4.河南理工大学 电气工程与自动化学院，河南 焦作 454000
摘要:综采工作面煤岩分界面识别是采煤机滚筒高度自适应调节的关键和难点，为了在不增加额外设备的情况下准确识别综采工作面煤岩分界面，从采煤机滚筒分别截割煤层和岩层的表现性状出发，提出一种基于改进RBF神经网络的采煤机截割煤岩性状智能识别方法，使采煤机滚筒能够高速实时判别煤岩。该方法根据采煤机截割电流、牵引电流和摇臂调高液压缸阻力的变化，采用改进的萤火虫算法对RBF神经网络的基函数参数进行优化，并采用优化后的 RBF神经网络模型对当前截割的煤岩性状进行识别。在耿村煤矿12150综采工作面实测数据的基础上开展试验，结果表明，基于改进RBF神经网络的煤岩性状识别模型对采煤机截割对象的识别准确率达到93.94％。利用该模型进行煤岩性状识别，无需加装额外探测设备，响应速度快、识别率高，有较好的工程应用潜力。
关键词:煤岩性状识别;采煤机;RBF神经网络;萤火虫算法
doi:10.16186/j.cnki.1673-9787.2020070089
基金项目:国家重点研发计划项目（2018YFC0604500）；河南省科技攻关项目（192102210100 ）；河南省高等学校重点科研项目 （19A413008）
收稿日期:2020/07/28
修回日期:2020/09/28
出版日期:2022/01/01
Intelligent recognition of coal and rock properties in shearer cutting process based on improved RBF neural network
DUAN Mingyu 1,2, YUAN Ruifu 3, YANG Yi 4
1.School of Mechenical Science and Engineering， Huazhong University of Science and Technology， Wuhan 430070 ， Hubei， China；2.Henan Dayou Energy Co. ，Ltd. ，Sanmen%ia 472300 ， Henan， China；3.School of Energy Science and Engineering， Henan Polytechnic University， Jiaozuo 454000 ， Henan， China；4.School of Electrical Engineering and Automation， Henan Polytechnic University， Jiaozuo 454000 ， Henan， China
Abstract: Recognition of the boundary between the coal and rock is the key issue of adjusting the drum high of the shearer． In order to recognize the boundary exactly without any other equipment，an intelligent method based on improved RBF neural network to identify the boundary according the different properties during the shearer cutting the coal and rocks． In this method，the different properties were reflect by the cutting current， traction current and the resistance of the hydraulic cylinder adjusting the height of the arm． Hence，the improved RBF neural network was used to analyze the properties，in which the parameters of the basis function inRBF neural network were optimized by the modified firefly algorithm． The verified experiments were carried out on the real data coming from 12150 workspace of Gengcun coal mine，and the experiment results showed that the recognition accuracy of the coal and rock property recognition model based on the improved RBF neural network proposed in this paper reached 93.94% ． The method described in this paper could be used to identify coal and rock properties without additional detection equipment． It had high response speed and recognition rate，and had great engineering application potential.
Key words:recognition of coal and rock property;shearer;RBF neural network;firefly algorithm
